English
Related papers

Related papers: Homogenized Transformers

200 papers

A many-particle theory is presented for the effective quasistatic permittivity of macroscopically homogeneous and isotropic systems of inhomogeneous dielectric particles with different degrees of penetrability. The theory is based upon our…

Soft Condensed Matter · Physics 2018-11-06 M. Ya. Sushko , A. V. Dorosh

We study the collective dynamics of a population of particles/organisms subject to self-consistent attraction-repulsion interactions and an external velocity field. The starting point of our analysis is a mean-field kinetic model and we…

Analysis of PDEs · Mathematics 2025-11-04 Thierry Goudon , Antoine Mellet

We introduce a stochastic generalisation of the classical deterministic Floquet-East model, a discrete circuit with the same kinetic constraint as the East model of glasses. We prove exactly that, in the limit of long time and large size,…

Statistical Mechanics · Physics 2025-01-29 Cecilia De Fazio , Juan P. Garrahan , Katja Klobas

Homogenisation theory has seen recent applications in deriving stochastic transport models for fluid dynamics. In this work, we first derive the stochastic Lagrange-to-Euler map that underpins stochastic transport noise in fluid dynamics as…

Mathematical Physics · Physics 2025-11-06 Theo Diamantakis , Ruiao Hu , James-Michael Leahy

This paper investigates the parabolic scaling limit of a damped stochastic wave map from the real line into the two-dimensional sphere, perturbed by multiplicative Gaussian noise of co-normal type. We prove that under this rescaling, the…

Probability · Mathematics 2025-07-29 Sandra Cerrai , Mengzi Xie

Asymptotic homogenisation is considered for problems with integral constraints imposed on a slowly-varying microstructure; an insulator with an array of perfectly dielectric inclusions of slowly varying size serves as a paradigm. Although…

Classical Analysis and ODEs · Mathematics 2023-04-03 A. Kent , S. L. Waters , J. Oliver , S. J. Chapman

Homogeneous nucleation is formulated within the context of fluctuating hydrodynamics. It is shown that for a colloidal or macromolecular system in the strong damping limit the most likely path for nucleation can be determined by gradient…

Soft Condensed Matter · Physics 2011-10-27 James F. Lutsko

We analyze the macroscopic behavior of multi-populations randomly connected neural networks with interaction delays. Similar to cases occurring in spin glasses, we show that the sequences of empirical measures satisfy a large deviation…

Mathematical Physics · Physics 2015-06-15 Tanguy Cabana , Jonathan Touboul

High-order Graph Neural Networks (HO-GNNs) have been developed to infer consistent latent spaces in the heterophilic regime, where the label distribution is not correlated with the graph structure. However, most of the existing HO-GNNs are…

Machine Learning · Computer Science 2023-06-30 Ahmed Begga , Francisco Escolano , Miguel Angel Lozano , Edwin R. Hancock

Although transformer-based models have shown exceptional empirical performance, the fundamental principles governing their training dynamics are inadequately characterized beyond configuration-specific studies. Inspired by empirical…

Machine Learning · Computer Science 2025-10-09 Zheng-An Chen , Tao Luo

We consider a stochastic many-body system where a source refills uniformly the empty sites of a hypercubic lattice, on which each particle is allowed to jump (symmetrically) onto neighboring vacant sites. In addition, there is a local {\it…

Statistical Mechanics · Physics 2007-05-23 Mauro Mobilia

Motivated by the recent contribution \cite{BB17} we study the scaling limit behavior of a class of one-dimensional stochastic differential equations which has a unique attracting point subject to a small additional repulsive perturbation.…

Mathematical Physics · Physics 2019-06-26 Martin Kolb , Matthias Liesenfeld

The existence of phase-separated states is an essential feature of infinite-volume systems with a thermal, first-order phase transition. At energies between those at which the phase transition takes place, equilibrium homogeneous states are…

In homogenization theory, mathematical models at the macro level are constructed based on the solution of auxiliary cell problems at the micro level within a single periodicity cell. These problems are formulated using asymptotic expansions…

Numerical Analysis · Mathematics 2025-06-10 P. N. Vabishchevich

We investigate dynamical heterogeneities in the collective relaxation of a concentrated microgel system, for which the packing fraction can be conveniently varied by changing the temperature. The packing fraction dependent mechanical…

Soft Condensed Matter · Physics 2011-05-31 David A. Sessoms , Irmgard Bischofberger , Luca Cipelletti , Véronique Trappe

We investigate the behavior of self-propelled particles in infinite space dimensions by comparing two powerful approaches in many-body dynamics: the Fokker-Planck equation and dynamical mean-field theory. The dynamics of the particles at…

The most frequently used in physical application diffusive (based on the Fokker-Planck equation) model leans upon the assumption of small jumps of a macroscopic variable for each given realization of the stochastic process. This imposes…

Statistical Mechanics · Physics 2007-05-23 Serge Shpyrko , V. V. Ryazanov

In this paper, we frame homogeneous-feature multi-task learning (MTL) as a hierarchical representation learning problem, with one task-agnostic and multiple task-specific latent representations. Drawing inspiration from the information…

Machine Learning · Computer Science 2022-10-04 João Machado de Freitas , Sebastian Berg , Bernhard C. Geiger , Manfred Mücke

Diffusion behaviors of heterogeneous materials are of paramount importance in many engineering problems. Numerical models that take into account the internal structure of such materials are robust but computationally very expensive. This…

Numerical Analysis · Mathematics 2023-12-18 Jan Eliáš , Hao Yin , Gianluca Cusatis

We develop a model in which interactions between nodes of a dynamic network are counted by non homogeneous Poisson processes. In a block modelling perspective, nodes belong to hidden clusters (whose number is unknown) and the intensity…

Machine Learning · Statistics 2017-07-11 Marco Corneli , Pierre Latouche , Fabrice Rossi
‹ Prev 1 4 5 6 7 8 10 Next ›